Research On Denoising And First Arrival Picking Of Microseismic Signal Under Complex Noise Background Of Coal Mine

A large number of microseismic events will occur in the process of rock fracture in underground coal mines.By recording and analyzing microseismic signals,the microseismic monitoring system can obtain numerous information of rock fracture,which plays a positive role in the safe mining of mines.However,due to the complex working environment in the well,many equipment work will generate a lot of noise.The microseismic signals collected by the pollution detector have a greater impact on the initial pick-up of microseismic events.The accuracy of the first arrival of the microseismic signal directly affects the accurate location of the microseismic event.Therefore,in order to improve the positioning accuracy,this thesis studies the noise reduction method and first break picking method of microseismic signal.First,the complex environment of coal mines is analyzed,and the problems of several noise reduction methods for microseismic signals are explained.A novel method for noise reduction based on adaptive ensemble empirical mode decomposition(NAEEMD)and frequency-constrained singular value decomposition(SVD)is proposed.The method firstly decomposes the noise-containing microseismic signals into multiple intrinsic mode functions(IMF)with order from high to bottom by NAEEMD method.With the signal-led IMF component as the guide,SVD is used to decompose and extract useful signals from the noise-led IMF component and the transition component.Finally,the IMF component decomposed by SVD is reconstructed,that is,the microseismic signal after denoising.Experimental results show that the SNR of the proposed method is more than 5d B higher than that of other methods,and the energy percentage of the signal is as high as 94.8625%,the signal standard deviation reached 0.0216,and all the quantitative evaluation indexes were better than other noise reduction methods.Then,in view of the weak anti-interference ability of the commonly used P wave arrival picking method and the poor picking effect.Based on the traditional method,an improved combined method of STA/LTA and Akakike Information Criteria(AIC)is proposed.This method first uses the improved STA/LTA method to determine the approximate location of the microseismic signal first arrival point,then selects the time window containing the microseismic signal,and finally uses the AIC method to determine the precise location of the microseismic signal first arrival point.The attenuated sinusoidal signal and the measured signal are used for verification.Compared with the results of STA/LTA and AIC methods,at the same time,a section of coal mine strong noise microseismic signal is selected for experiment.Experimental results show that this method is more accurate in picking up the first arrival points of microseismic signals,and it can also pick up the first arrival points of microseismic signals under complex noises of coal mines.Finally,a set of microseismic signal acquisition and processing system is designed.The system consists of data acquisition part and human-computer interaction part.The data acquisition part is based on the main control chip of STM32F107,which realizes the acquisition,processing,storage and transmission of the signal.The human-computer interaction part is built on the Lab VIEW platform,which realizes the signal reception,display,storage,playback,signal noise reduction and first arrival point picking.The system test results show that it can collect microseismic signals with a wide dynamic range and display them in real-time as waveforms,and can perform noise reduction and P-wave first arrival pick-up processing on the collected signals.The development of the system lays a foundation for the accurate positioning of microseismic events.